A systematic review of circulating markers in primary chronic venous insufficiency

Abstract

Objectives

The etiology of primary chronic venous insufficiency is poorly understood. This systematic review aimed to summarize published evidence assessing the association of circulating markers with primary chronic venous insufficiency.

Methods

A search was undertaken through the PubMed database using the terms “venous insufficiency” and “biological marker” or “plasma” or “serum”. Search limits included English language, human subjects and studies with publication dates from 1994. Studies which classified patients using the Clinical-Etiology-Anatomy-Pathophysiology system of venous disease were analyzed.

Results

Seventeen studies were included, which have examined > 60 different biomarkers. A total of 13 markers were assessed in >1 study with the number of primary chronic venous insufficiency cases ranging from 41 to 244 and the number of controls ranging from 30 to 144 in these studies. Circulating estradiol, homocysteine and vascular endothelial growth factor were the most consistently associated with primary chronic venous insufficiency.

Conclusions

Whilst a number of studies have examined biomarkers associated with primary chronic venous insufficiency, further studies are required using improved and standardized approaches on larger populations. Biomarker research may increase pathogenic knowledge and result in opportunities to decrease chronic venous insufficiency burden.

Keywords

Venous insufficiency biological markers blood serum plasma

Introduction

Chronic venous insufficiency (CVI) is one of the most common vascular diseases in the developed world and is a major contributor to psychosocial morbidity. As much as 1.5–2% of the annual healthcare budgets in European countries are spent on the management of CVI.^1,2 The most severe manifestation of disease, venous leg ulceration, is present in up to 1.5% of the population and in subjects 65 years and older, this figure increases to 4%.^3,4 Whilst treatment with compression bandaging remains the cornerstone of therapy, management deficiencies exist with as many as 50% of ulcers remaining active for more than 1 year.^2,5 A clearer understanding of CVI pathogenesis could provide a means to improve current management deficiencies.

Current theories regarding the pathogenesis of CVI include the pericapillary fibrin cuff and fibrinolytic abnormalities hypotheses; the growth factor “trap” hypothesis and the white cell hypothesis.^6–10 Circulating biomarker studies have been identified as having potential to improve pathogenic understanding, increase prognostic information, enhance diagnostic abilities as well as provide treatment targets.^9,11 This review aimed to (1) critically appraise published studies that have assessed the association of circulating markers with primary CVI and (2) to further assess those markers that were examined for association with primary CVI in at least two studies.

Methods

Protocol and focus

This systematic review was performed with a standard written protocol that followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.¹² We sought studies in which biomarkers were assessed in the circulating blood of subjects with primary CVI and in which patients were classified according to the CEAP classification system. This classification includes information on clinical manifestations (C), etiological factors (E), anatomical distribution of disease (A) and underlying pathophysiological findings (P).¹³

Search criteria

A systematic approach was undertaken to identify relevant literature by searches performed through the PubMed database. The search terms “venous insufficiency” and “biological marker” or “plasma” or “serum” were used. Titles and abstracts of studies identified were reviewed for their appropriateness and studies of possible significance were read fully and further subjected to a number of inclusion and exclusion criteria. To increase the yield of relevant publications, the reference lists of those studies initially identified were also searched.

Eligibility criteria

In order to be included in this review, studies needed to be conducted on human subjects and report their findings in English. Studies published between 1 January 1994 and 1 April 2013 were examined, focusing on the period after the first CEAP consensus report. Included studies needed to measure and report the circulating concentrations of markers in the blood of patients with primary CVI. The included studies needed to have classified patients using at least the clinical and etiology portion of the CEAP system. Each study was required to have ≥10 subjects in the groups compared. Those studies that investigated secondary or congenital CVI were excluded, as were reports of alterations in the levels of circulating markers in response to different treatment options.

Data collection process

One author extracted the data by means of recording patient characteristics, procedural details and biomarker associations with disease. Patient characteristics documented included number of cases, gender and age as well as the laboratory investigations undertaken. Procedural details included the site and methods of blood collection. Any marker assessed in >1 study was evaluated further and tabulated including number of cases and controls, clinical classification of cases, techniques used to assess the marker and any significant associations. Of these markers, those assessed in ≥100 subjects were analyzed further.

Data assessment

Tables were generated to enable comparisons. Considerable variation in the data reported was identified and as such the data format was standardized where possible. Further analysis was conducted to assess the overall quality of studies identified. No approved quality assessment tool exists for biomarker studies. We therefore adapted criteria validated for quality assessment of retrospective studies and generated a specific tool that took into account three key features including study information, biomarker information and statistical analysis.¹⁴ The ultimate aim was to report an overall quality assessment score.

Results

Study selection

A total of 401 articles were identified via searching PubMed (n = 392) and reference lists of relevant publications (n = 9), as summarized in Figure 1. In all, 54 articles were excluded since they were duplicates. Following review of titles and abstracts 312 studies were excluded due to failure to focus on circulating markers in CVI. Full-text articles were then assessed (n = 35), leading to further exclusions. Reasons included: failure to use the CEAP classification system (n = 6); failure to investigate ≥10 primary CVI patients (n = 7); investigation of fluid wound exudate or tissue biopsy samples (n = 4) and not reporting concentrations of the circulating markers measured (n = 1). In total, 17 studies were identified as meeting inclusion and exclusion criteria for this review.

Figure 1.

Modified Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram to illustrate the studies identified for this review.

Study characteristics

Of the 17 studies identified, 14 were case-control studies, including a total of 622 cases and 644 controls,^15–28 and the remaining studies related the concentration of markers to the severity of venous disease in a total of 212 subjects (Table 1).^29–31 The total sample size of the studies identified varied considerably. Six studies included <50 patients^{20–22,24,26,30}; 8 studies included between 50 and 100 patients^{15,17,18,23,25,27,29,31} and 3 studies included >100 patients.^16,19,28 Many studies investigated a large number of biomarkers. One study reported the concentrations of 14 circulating factors,²³ whilst only 4 studies investigated 1 circulating factor.^17–19,28

Table 1.

Summary of biomarker studies.

Reference	Biomarkers	Recruitment method	Cases, n	Controls, n
15^a	Iron, malonyldialdehyde, total antioxidant capacity, uric acid	Patients considered for varicose vein surgery	35	23
16^a	Anti-phospholipid antibodies, anti-thrombin deficiencies, factor VIII, factor IX, factor XI, homocysteine, protein C, protein S	Consecutive patients referred with uncomplicated CVI	54	54
17^a	Homocysteine	Not stated	65	35
18^a	NO	Patients from Vascular outpatient clinic	44	13
19^a	VEGF	Patients from Vascular outpatient clinic	108	30
20^a	Androstendione, dehydroepiandrostendion, estradiol, testosterone	Patients from clinic	21	13
21^a	Androstendione, estradiol, haemoglobin, leukocytes, platelets, testosterone	Patients from Department of Dermatology, Venerology and Allergology	20	20
22^a	Platelet activation status (mean channel fluorescence), platelet count, soluble P-selectin	Not stated	21	13
23^a	Activities of lysosomal enzymes (elastase, lysozyme, myeloperoxidase, Beta-D-glucuronidase), CD11a (β2-integrin α-chain), CD11b (β2-integrin α-chain), CD18 (β2-integrin β-chain), CD15 (Lewis' antigen X), CD31 (platelet-endothelial adhesion molecule-1), CD54 (intercellular adhesion molecule-1), CD49d (α-chain of very late antigens-4), CD62L (L-selectin), INT reduction, superoxide	Not stated	26	39
24^a	Granulocyte pseudopod formation, hydrogen peroxide production, nitroblue tetrazolium reduction	Not stated	21	16
25^a	Fibrinogen, IL-6, IL-12, malonyldialdehyde, protein C, protein S, tissue plasminogen activator, VEGF, von Willebrand factor	Patients attending hospital for management of venous disease	25	25
26^a	% transferrin iron binding capacity, ferritin, iron, MMP-9 (total, activated MMP-9, rate of activated MMP-9), transferring	Not stated	30	14
27^a	RBC CD35 expression, RBC Fy6 expression, IL-8	Not stated	43	8
28^a	Lipoprotein (a)	Not stated	109	341
29^b	Estradiol, glucose, insulin, sex hormone binding globulin, testosterone	Patients recruited from newspaper and broadcasting advertisement	90	N/A
30^b	Angiotensin-converting enzyme activity, IL-8, lactoferrin, L-selectin, MMPs 9 and 2, myeloperoxidase, PVO₂, sICAM-1, sVCAM-1, thrombomodulin, VEGF, von Willebrand factor	Not stated	22	N/A
31^b	Folate, homocysteine, vitamin B12	Patients from hospital-based clinic	100	N/A

CD: cluster differentiation; CVI: chronic venous insufficiency; IL: interleukin; INT: 3-(4-iodophenyl)-2-(4-nitrophenyl)-5-phenyltetrazolium chloride; MMP: matrix metalloproteinase; N/A: not applicable; NO: nitric oxide; PVO₂: partial pressure of oxygen in venous blood; RBC: red blood cell; sICAM: soluble intercellular adhesion molecule; sVCAM: soluble vascular cell adhesion molecule; VEGF: vascular endothelial growth factor.

Case-control biomarker studies.

Studies on venous patients with no controls.

The general characteristics of the patients included in each of the studies are shown in Supplementary Table 1. The mean age ranged from 36 to 72 years in CVI cases and from 33 to 54 years in controls. Of the 16 studies reporting gender data, males comprised 329 of the 834 (39%) CVI cases.^15–26,28 Additionally, 308 of 644 (48%) controls were noted to be males in the 13 studies reporting this data.^15–26,28 CVI was confirmed in cases using a variety of imaging techniques including duplex ultrasonography (n = 14),^{15–22,24–28,30} photoplethysmography (n = 3),^18,19,23 venography (n = 2),^22,27 air plethysmography (n = 1)²⁶ and venous strain-gauge plethysmography (n = 1).²⁹ In addition, five studies excluded arterial disease by measurement of ankle pressures using Doppler.^{16,17,23,28,31} In regard to the control subjects, CVI was excluded by physical examination in six studies.^{15,17–19,24,25} One study used a control group of 341 healthy individuals previously recruited by other investigators.²⁸ Other studies excluded the presence of CVI in controls using duplex ultrasonography (n = 6),^{16,20–22,26,27} photoplethysmography (n = 1)²³ and air plethysmography (n = 1).²⁶ One study also performed measurement of ankle pressures using Doppler in controls.¹⁶

Use of the quality assessment tool illustrated differences between the studies. In those studies which compared cases and controls (Supplementary Table 2), most studies did not assess controls by duplex ultrasonography nor adjust for key confounders. In those studies in which >2 circulating markers were investigated, no study considered the possibility of multiple testing. Three studies compared biomarkers in venous patients (Supplementary Table 3). Of these studies, only 1 assessed subjects by duplex ultrasonography and few key confounders were considered.

Assessment techniques

One of the greatest differences identified between the studies was the variation in the site and method of blood collection (Supplementary Table 4). A proportion of studies collected blood from the leg affected by venous disease. Sites of collection included the ankle vein, dorsal foot vein, long saphenous vein and varicose veins with 7 of the 17 included studies reporting taking blood from one of these sites.^{18,19,21,22,26,27,30} The procedural details varied with a small number of studies taking blood with the subjects lying supine.^{18,19,22,24,30} Other studies had patients stand or sit for a period of time in order to induce experimental venous hypertension in the leg.^{21,22,26,27,30} A number of other studies collected blood from the arm of the patients investigated.^{15,16,20,21,24,26,30,31} Of the 17 studies, 4 did not state the site of blood collection.^17,23,29,30 One study took the interesting approach of taking blood from the varicose vein as a case sample and blood from the brachial vein of the same patient as a control sample.³⁰

After collection of blood, investigators then used the sample to measure the circulating factor or factors of interest in whole blood, serum or plasma. In those markers measured in >1 study, all investigations were conducted on the same component of blood, except in the cases of homocysteine, matrix metalloproteinase-9 (MMP-9) and myeloperoxidase. Homocysteine was measured in serum in one study²⁵ and plasma in three studies.^16,17,31 Different forms of MMP-9 were measured in plasma³⁰ and serum²⁶ in the two studies measuring this marker. Finally, myeloperoxidase was measured in plasma in one study³⁰ and determined in peripheral neutrophils in another.²³

Association of circulating biomarkers with CVI

Of the 64 different biomarkers that have been assessed, 51 were examined in only one study. Despite some studies finding a positive association between markers and severity or presence of disease, it is difficult to comment whether a consistent association exists. As such, markers assessed in >1 studies were examined including androstendione,^20,21 homocysteine,^16,17,25,31 interleukin-8,^27,30 iron,^15,26 malonyldialdehyde,^15,25 matrix metalloproteinase-9,^26,30 myeloperoxidase,^23,30 nitric oxide,^18,25 estradiol,^20,21,29 protein C,^16,25 testosterone,^20,21,29 vascular endothelial growth factor (VEGF)^19,25,30 and von Willebrand factor.^25,30 Of these, only four markers were assessed in ≥100 CVI cases.

Homocysteine

Homocysteine represents the most widely investigated circulating marker in CVI. Four independent investigations examined homocysteine concentrations in 244 CVI patients (Table 2).^16,17,25,31 The most consistently reported finding was that CVI cases had a raised concentration when compared to controls and the circulating concentration was highest in more severe stages of disease. Of the three studies reporting this association, the largest undertaken by Sam and colleagues³¹ found that hyperhomocysteinemia (≥15 µmol/L) was significantly related to clinical grade in the 100 patients they investigated (p = 0.001). Twenty-three percent of patients with varicose veins, 20% of patients with edema, 39% of patients with venous skin changes, 53% of patients with a healed venous ulcer and 65% of patients with active venous ulceration met their criteria of having hyperhomocysteinemia.³¹

Table 2.

Association of circulating homocysteine with chronic venous insufficiency (CVI).

Reference	Cases, n	Clinical (C) classification	Concentration, cases	Controls, n	Concentration, controls	Method	Significant associations
16^a,b	54	C2–C3 (n = 27)	≥16.6; n = 4	27^c	≥16.6; n = 3	High-pressure liquid chromatography	HHcy^d more common in C5–6 than controls (40.7% vs. 7.4%; OR, 5.50; 95CI, 1.20–51.07). Similar in C2–C3 and controls (16.0% vs. 12.0%; OR, 2.00; 95% CI, 0.43–12.36)
16^a,b	54	C5–6 (n = 27)	≥16.6; n = 11	27^c	≥16.6; n = 2	High-pressure liquid chromatography
17^a,b,e	65	C4 (n = 25)	19.1 (15–28)	35	8.1 (5–12.2)	Fluorescence polarization immunoassay	Cases C4-controls (p < 0.00); C6-controls (p < 0.00). No significant difference C4–C6 (p = 0.877)
17^a,b,e	65	C6 (n = 40)	18.98 (15–29)	35	8.1 (5–12.2)	Fluorescence polarization immunoassay
25^b,f	25	C2 (n = 25)	12.7 ± 2.8	25	14.9 ± 7.3	Chemiluminescence immunoassay	NS
31^a,e,g	100	C2 (n = 39)	11.6 (9.2–14.0)	N/A	N/A	High-pressure liquid chromatography	HHcy^h 39% (p < 0.001, binomial test vs. normal population). Significantly related (Pearson χ2 for trend, 13.616; p < 0.009) to clinical grade: C2, 23%; C3, 20%; C4, 39%; C5, 53%; C6, 65%.
		C3 (n = 10)	11.5 (9.9–13.1)
		C4 (n = 13)	12.5 (10.7–16.0)
		C5 (n = 15)	15.1 (11.1–16.0)
		C6 (n = 23)	18.1 (13.7–22.4)

Clinical (C) classification: C2: varicose veins, C3: edema, C4: skin changes; C5: healed venous ulcer, C6: active venous ulcer; ELISA: enzyme-linked immunosorbent assay; HHcy: hyperhomocysteinemia; NS: not significant.

Measured in plasma.

Units given as µmol/L.

Age- and sex-matched controls.

Hyperhomocysteinemia defined as ≥16.6 µmol/L.

Numbers given as median (interquartile range).

Measured in serum; numbers are mean ± standard error of mean.

Units given as µpmol/L.

Hyperhomocysteinemia defined as ≥15 µmol/L.

Darvall et al.¹⁶ also found a similar association despite defining hyperhomocysteinemia (≥16.6 µmol/L) differently. Patients with stasis dermatitis (C4) and active ulceration (C6) had a significantly higher level of homocysteine when compared to age- and sex-matched controls.¹⁶ However, the authors reported similar prevalence of hyperhomocysteinemia when comparing the varicose vein and edema cases to matched controls at 16% compared to 12%, respectively.¹⁶ Another study by Durmazlar et al.¹⁷ found a statistically significant increased level of homocysteine in C4 (n = 25) and C6 (n = 40) cases compared to controls (n = 35) (p < 0.05).

Only one study reported no difference in homocysteine concentrations between case and control subjects.²⁵ Notably, this was the smallest study, that measured homocysteine in serum and specifically focused on differences between early stage venous disease and controls.²⁵

Vascular endothelial growth factor

Plasma levels of VEGF have been measured in three studies with two studies subsequently reporting higher VEGF levels in cases compared to controls (Table 3).^19,25,30 Whilst both Howlader and Coleridge Smith¹⁹ and Yasim et al.²⁵ found higher concentrations in CVI cases, Howlader and Coleridge Smith¹⁹ reported that statistical significance was only reached when the C5 group in their study was included in the analysis. This was despite investigating a total of 108 cases of which 47 were classified as C2.¹⁹ Yasim et al.²⁵ reported statistically significant increased VEGF in cases compared to controls and notably all cases were clinically classified as having C2 disease. The smallest study by Jacob et al.³⁰ found no significant differences in 22 subjects but in contrast to the other studies used blood from the arm as the control sample and blood from the leg as the case sample.

Table 3.

Association of circulating vascular endothelial growth factor with chronic venous insufficiency (CVI).

Reference	Cases, n	Clinical (C) classification	Concentration, cases, pg/mL	Controls, n	Concentration, controls, pg/mL	Method	Significant associations
19^a	108	C2 (n = 47)	85 (49–127)	30	56 (38–85)	ELISA	Higher in cases than controls, but statistical significance only reached in C5 group compared to controls. Levels were ∼60% higher for cases than controls
		C3 (n = 16)
		C4 (n = 29)
		C5 (n = 16)
25^b	25	C2 (n = 25)	258.5 ± 139.7	25	49.5 ± 23.7	ELISA	Higher in cases than controls (p < 0.01)
30^c	22^d	C3 (n = 12)	At rest: 48.8 ± 42.2.9; After blood stasis: 41.7.6 ± 28.2	22^e	At rest: 44.3 ± 60.9; After blood stasis: 33.6 ± 14.1; At rest: 34.0 ± 19.3	Immunoassay	NS
30^c	22^d	C4 (n = 10)	At rest: 48.8 ± 42.2.9; After blood stasis: 41.7.6 ± 28.2	22^e		Immunoassay	NS

Clinical (C) classification: C2: varicose veins, C3: edema, C4: skin changes; C5: healed venous ulcer; ELISA: enzyme-linked immunosorbent assay; NS: not significant.

Numbers given as median (interquartile range).

Numbers given as mean ± standard error of mean.

Numbers given as mean ± standard deviation.

Same subjects used for case and control: blood from varicose vein (case sample).

Same subjects used for case and control: blood from brachial vein (control sample).

Estradiol

Three studies have examined the circulating levels of estradiol in a cumulative study population of 131 cases and 33 controls; however, variation in approaches and the gender of study cohorts was evident (Table 4).^20,21,29 Ciardullo et al.²⁹ measured estradiol in 90 post-menopausal women after an overnight fast and after adjusting for body mass index (BMI), fasting glucose and insulin values as covariates. The authors found a higher circulating level of estradiol was associated with the clinical presentation of varicose veins.²⁹

Table 4.

Association of circulating estradiol with chronic venous insufficiency (CVI).

Reference	Cases, n	Clinical (C) classification	Concentration, cases	Controls, n	Concentration, controls	Method	Significant associations
20^a	21	C3 (n = 17)	110.97 ± 52.09	13^b	89.12 ± 17.02	Estradiol-specific antibody and ruthenium- labelled estradiol peptide	NS
20^a	21	C4 (n = 4)	110.97 ± 52.09	13^b	89.12 ± 17.02		NS
21^a	20	C2 (n = 2)	Arm: 93.2 ± 28.5; Leg: 242.2 ± 208.6	20^c	Arm: 88.2 ± 32.7; Leg: 133.2 ± 42.9	Immunoassay	Cases-leg higher levels (242.2 [79–941]) compared to cases-arm (93.2 [31–147]) (Arm/leg difference: p < 0.001). Control-arm and control-legs were significantly different (p < 0.001)
		C3 (n = 9)
		C4 (n = 7)
		C5 (n = 1)
		C6 (n = 1)
29^d	90	C0, C1 (n = 62)	8.7 ± 3.7	N/A	N/A	Radioimmunoassay	Estradiol in upper tertile of frequency distribution associated with varicose veins (prevalence odds ratios: 3.6; 95% CI 1.1–11.6) and with increased lower limb venous distensibility (prevalence odds ratios 4.4; 95% CI 1.2–15.5)
29^d	90	C2, C3 (n = 28)	8.7 ± 3.7	N/A	N/A	Radioimmunoassay

Clinical (C) classification: C0: no visible or palpable signs; C1: telangiectasies or reticular veins; C2: varicose veins, C3: edema, C4: skin changes; C5: healed venous ulcer, C6: active venous ulcer. Numbers are mean ± standard deviation.

Units given as pmol/L.

Control group healthy men C ≤ 1.

Controls: C0 (n = 12); C1 (n = 8).

Units given as pg/mL.

The other identified studies investigated only male patients. The study by Kendler et al.²⁰ found no statistical difference in the serum estradiol levels of blood taken from the antecubital vein of 21 men with symptomatic varicose veins compared to 13 healthy men. The remaining study examined blood taken from both the leg and arm of cases and controls and found a significant difference in the mean serum levels of estradiol of the cases and controls when comparing the arm and leg samples (arm/leg difference in cases and control, p < 0.001).²¹ A statistically significant difference was also noted in the levels of estradiol in the case leg and control leg samples (p = 0.021).²¹ Notably, in both of these studies focusing on males, the control groups were defined as having a clinical classification of C ≤ 1, meaning that a proportion of the controls had telangiectasies or reticular veins.^20,21

Testosterone

The three studies investigating estradiol also measured testosterone (Table 5).^20,21,29 No statistically significant differences in testosterone were observed in two of the studies,^20,29 but the remaining investigation by Kendler et al.²¹ found a significant difference between arm and leg mean testosterone levels (arm/leg difference, p < 0.001) as well as an increased testosterone concentration in the leg of cases when compared to the leg of controls (p = 0.04). One study calculated an estradiol and free testosterone ratio in blood taken from the antecubital vein and found that in 21 cases the ratio was significantly higher than the 13 controls (CVI cases: 2.83 ± 0.79 vs. controls: 2.32 ± 0.63, p < 0.05).²⁰

Table 5.

Association of circulating testosterone with chronic venous insufficiency (CVI).

Reference	Cases, n	Clinical (C) classification	Concentration, cases	Controls, n	Concentration, controls	Method	Significant associations
20^a	21	C3 (n = 17)	40.11 ± 13.32	13^b	40.27 ± 9.62	Competitive radioimmunoassay	NS
20^a	21	C4 (n = 4)	40.11 ± 13.32	13^b	40.27 ± 9.62	Competitive radioimmunoassay	NS
21^c	20	C2 (n = 2)	Arm: 15.5 ± 5.0; Leg: 44.9 ± 61.1	20^d	Arm: 16.1 ± 6.0; Leg: 18.0 ± 10.0	Immunoassay	Cases-arm and cases-leg significant difference (arm/leg difference, p < 0.001). Higher in cases-leg compared to controls-leg (p = 0.04)
		C3 (n = 9)
		C4 (n = 7)
		C5 (n = 1)
		C6 (n = 1)
29^e	90	C0, C1 (n = 62)	0.43 ± 0.12	N/A	N/A	Radioimmunoassay	NS
29^e	90	C2, C3 (n = 28)	0.43 ± 0.12	N/A	N/A	Radioimmunoassay	NS

Units give as pmol/L.

Controls defined as C ≤ 1.

Units given as nmol/L.

Controls: C0 (n = 12); C1 (n = 8).

Units given as ng/mL.

Discussion

The main finding of this review is that only a limited number of biomarkers have been investigated for association with CVI in multiple studies. Currently, circulating estradiol, homocysteine and VEGF are the biomarkers most consistently associated with primary CVI. Each of these biomarkers may play a specific role in pathogenic steps involved in venous ulceration development. Homocysteine has been recognized as increasing endothelial-leukocyte interaction³² as well as venous thromboembolism development.³³ Estradiol and VEGF have been found to have a role in the coagulation and inflammatory responses and overall vascular tone.^6,9

It appears that the pathogenesis of primary CVI involves multiple pathways leading to differential expression of numerous biomarkers. For instance, an increased concentration of homocysteine in the lower leg combined with venous hypertension and subsequent blood stasis may increase leukocyte interaction with the endothelium and this may ultimately lead to leukocyte trapping.³² This adds support for the white cell trapping hypothesis resulting in vascular damage and increased vascular permeability.^6,34–36 High circulating homocysteine is also associated with increased risk of thrombosis and this in-turn may mean that the pericapillary fibrin cuff and fibrinolytic abnormalities hypotheses are supported, with the development of microembolisms thereby altering oxygen and nutrients exchange in the lower leg.^25,33 It is likely that venous leg ulceration is also associated with alterations in multiple biomarkers involved in numerous underlying pathways. Current work suggests that the development of CVI is multifactorial and if a final biomarker panel was to be clinically useful it would need to include multiple unrelated and uncorrelated markers in order to maximize scientific value and clinical discrimination.

Critical analysis of the literature highlights the need for caution in interpretation of the currently available studies assessing the association of circulating markers with venous disease. It is important to acknowledge the general limitation of reverse causality inherent in biomarker research. Additionally, a large number of confounding factors affecting the biomarkers investigated are likely to be present in the study cohorts. Of the three studies investigating estradiol, only one adjusted for BMI, blood sugar levels and insulin values as covariables.²⁹ Indeed, any number of variables may affect the circulating levels of the markers identified and this point highlights the limited value of research conducted on small populations. Whilst it would be impossible to exclude all confounding factors, a possible solution may be to undertake large-scale studies and to adjust for confounders using multivariate analysis.

If large-scale studies were to be conducted, there is an obvious need for cases to be classified in order to improve comparison abilities. The CEAP classification system represents a useful means for doing so, but some studies cited applied this system to patients with limited laboratory evaluation. Currently, duplex ultrasonography is the most widely used assessment of the severity of disease and failure to utilise this evaluation may ultimately mean that secondary CVI cases are included in research efforts.^6,9,13,37 This is important to highlight as homocysteine has been associated with thrombophilia and studies that have not employed duplex ultrasonography may have investigated cases of secondary CVI rather than primary CVI.

It should also be noted that a significant number of studies only assessed controls by clinical examination. A proportion of the population is recognised as having venous insufficiency but no clinical signs.⁹ This point further emphasises the need for duplex ultrasonography to be utilised, even in control populations. It may mean that some studies may have overlooked biomarkers that may have a role in the pathogenesis of CVI. Studies using controls that were noted to have C1 venous disease may have also underreported biomarkers associations.

This review identified three biomarkers that have been found to have an association with CVI and all can be linked to alteration of vascular wall homeostasis. The implications of the findings suggest that CVI pathogenesis is likely to be multifactorial and a complex process. The most valuable outcome of biomarker research may be that findings assist practitioners by improving prognostic information on patients with CVI. This would allow the earlier initiation of preventative strategies with the potential of minimizing disease morbidity. Currently, however, there has been no assessment of whether any of the identified biomarkers can significantly aid clinical decision making for patients with primary CVI.

Footnotes

Acknowledgements

JG holds a Practitioner Fellowships from the National Health and Medical Research Council, Australia (1019921). JG holds a Senior Clinical Research Fellowship from the Office of Health and Medical Research.

Funding

Funding from the Office of Health and Medical Research, Queensland Government and National Health and Medical Research Council supported this work.

Conflict of Interest

None declared.

References

Ruckley

. Socioeconomic impact of chronic venous insufficiency and leg ulcers. Angiology 1997; 48: 67–69.

Simka

Majewski

. The social and economic burden of venous leg ulcers: focus on the role of micronized purified flavonoid fraction adjuvant therapy. Am J Clin Dermatol 2003; 4: 573–581.

Margolis

Bilker

Santanna

. Venous leg ulcer: incidence and prevalence in the elderly. Am J Clin Dermatol 2002; 46: 381–386.

Meissner

Gloviczki

Bergan

. Primary chronic venous disorders. J Vasc Surg 2007; 46: S54–S67.

Eberhardt

Raffetto

. Chronic venous insufficiency. Circulation 2005; 111: 2398–2409.

Abbade

LPF

Lastória

. Venous ulcer: epidemiology, physiopathology, diagnosis and treatment. Int J Dermatol 2005; 44: 449–456.

Pappas

Cerveira

Lal

. Causes of severe chronic venous insufficiency. Semin Vasc Surg 2005; 18: 30–35.

Raffetto

Khalil

. Mechanisms of varicose vein formation: valve dysfunction and wall dilation. Phlebology 2008; 23: 85–98.

Valencia

Falabella

Kirsner

. Chronic venous insufficiency and venous leg ulceration. J Am Acad Dermatol 2001; 44: 401–421.

10.

Golledge

Quigley

. Pathogenesis of varicose veins. Eur J Vasc Endovasc 2003; 25: 319–324.

11.

Moore

Huddleston

Stacey

. Venous leg ulcers ‐ the search for a prognostic indicator. Int Wound J 2007; 4: 163–172.

12.

Moher

Liberati

Tetzlaff

. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med 2009; 151: 264–269.

13.

Eklof

Rutherford

Bergan

. Revision of the CEAP classification for chronic venous disorders: Consensus statement. J Vasc Surg 2004; 40: 1248–1252.

14.

Downs

Black

. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health 1998; 52: 377–384.

15.

Budzyń

Iskra

Krasiński

. Serum iron concentration and plasma oxidant-antioxidant balance in patients with chronic venous insufficency. Med Sci Monit 2011; 17: CR719–CR727.

16.

Darvall

KAL

Sam

Adam

. Higher prevalence of thrombophilia in patients with varicose veins and venous ulcers than controls. J Vasc Surg 2009; 49: 1235–1241.

17.

Durmazlar

Akgul

Eskioglu

. Hyperhomocysteinemia in patients with stasis dermatitis and ulcer: A novel finding with important therapeutic implications. J Dermatolog Treat 2009; 20: 336–339.

18.

Howlader

Coleridge Smith

. Increased plasma total nitric oxide among patients with severe chronic venous disease. Int Angiol 2002; 21: 180–186.

19.

Howlader

Coleridge Smith

. Relationship of plasma vascular endothelial growth factor to CEAP clinical stage and symptoms in patients with chronic venous disease. Eur J Vasc Endovasc 2004; 27: 89–93.

20.

Kendler

Blendinger

Haas

. Elevated serum estradiol/testosterone ratio in men with primary varicose veins compared with a healthy control group. Angiology 2009; 60: 283–289.

21.

Kendler

Makrantonaki

Kratzsch

. Elevated sex steroid hormones in great saphenous veins in men. J Vasc Surg 2010; 51: 639–646.

22.

Chen

. Venous hypertension induces increased platelet reactivity and accumulation in patients with chronic venous insufficiency. Angiology 2006; 57: 321–329.

23.

Stvrtinova

Jahnova

Weissova

. Inflammatory mechanism involving neutrophils in chronic venous insufficiency of lower limbs. Bratisl Lek Listy 2001; 102: 235–239.

24.

Takase

Schmid-Schönbein

Bergan

. Leukocyte activation in patients with venous insufficiency. J Vasc Surg 1999; 30: 148–156.

25.

Yasim

Kilinc

Aral

. Serum concentration of procoagulant, endothelial and oxidative stress markers in early primary varicose veins. Phlebology 2008; 23: 15–20.

26.

Zamboni

Scapoli

Lanzara

. Serum iron and matrix metalloproteinase-9 variations in limbs affected by chronic venous disease and venous leg ulcers. Dermatol Surg 2005; 31: 644–649.

27.

Zhang

. Immune function of erythrocytes in patients with chronic venous insufficiency of the lower extremities. Chin Med J (Engl) 2007; 120: 2224–2224.

28.

Zutt

Krüger

Rosenberger

. Increased lipoprotein (a) concentrations in patients with chronic venous leg ulcers: a study on patients with or without postthrombotic syndrome. Wound Repair Regen 2011; 19: 168–172.

29.

Ciardullo

Panico

Bellati

. High endogenous estradiol is associated with increased venous distensibility and clinical evidence of varicose veins in menopausal women. J Vasc Surg 2000; 32: 544–549.

30.

Jacob

Cazaubon

Scemama

. Plasma matrix metalloproteinase-9 as a marker of blood stasis in varicose veins. Circulation 2002; 106: 535–538.

31.

Sam

Burns

Hobbs

. The prevalence of hyperhomocysteinemia, methylene tetrahydrofolate reductase C677T mutation, and vitamin B12 and folate deficiency in patients with chronic venous insufficiency. J Vasc Surg 2003; 38: 904–908.

32.

Dudman

Temple

Guo

. Homocysteine enhances neutrophil-endothelial interactions in both cultured human cells and rats In vivo. Circ Res 1999; 84: 409–409.

33.

Den Heijer

Koster

Blom

. Hyperhomocysteinemia as a risk factor for deep‐vein thrombosis. N Engl J Med 1996; 334: 759–762.

34.

Smith

. Neutrophil activation and mediators of inflammation in chronic venous insufficiency. J Vasc Res 1999; 36: 24–36.

35.

Smith

. The causes of skin damage and leg ulceration in chronic venous disease. Int J Low Extrem Wounds 2006; 5: 160–168.

36.

Smith

PDC

Thomas

Scurr

. Causes of venous ulceration: a new hypothesis. Br Med J (Clin Res Ed) 1988; 296: 1726–1726.

37.

Kistner

Eklof

Masuda

. Diagnosis of chronic venous disease of the lower extremities: the” CEAP” classification. Mayo Clin Proc 1996; 71: 338–345.